The present invention relates to resistive furnaces of the type employed for heating a sample positioned in a resistive crucible held between a pair of electrodes.
In resistive furnaces of the type commercially available from Leco Corporation of St. Joseph, Michigan, as part of an analyzing instrument Model No. 760-200, a pair of opposed electrodes is provided between which there is positioned an electrically resistive graphite crucible into which a specimen to be analyzed is inserted. One of the electrodes is moved by means of a spring-loaded arm which clamps the graphite crucible between the copper electrodes whereupon an electrical current is passed through the resistive crucible to raise the crucible temperature to approximately 2500.degree. Centigrade thereby fusing the sample. A passage is provided for a carrier gas such as helium used for sweeping the gases from the crucible to an analyzer to determine, for example, the nitrogen or oxygen content of the specimen as carbon monoxide.
In such apparatus, samples of approximately 0.1- 1.5 grams are employed and frequently a flux is employed to facilitate the fusion of the sample. Due to the inherent limitations of the melting point of the copper electrodes previously employed, the required time for analysis for the maximum current (either A.C. or D.C.) employed can be relatively lengthy and in the order of several seconds.
In order to more efficiently utilize such a furnace, several improvements have been made, one of which is represented by U.S. Pat. No. 3,636,229 issued Jan. 18, 1972 to the present assignee. This patent discloses an improved crucible construction to more effectively heat the specimen contained therein. Even with such crucible construction, however, the copper electrode construction inherently limits the speed of operation as well as the temperature which can be reached by the furnace. Also, the relatively soft copper electrodes tend to wear, particularly since even though they are water-cooled, the tips engaging the crucible tend to melt. If the tips get too hot, they react, tending to cause erroneous readings for the specimen being tested.
Additionally, with the spring-actuated electrodes, when variations in fabricated crucible lengths are employed, frequently the contact resistance and contact between the electrode and the crucible varies significantly with different clamping pressures tending to affect the reproducibility of the furnace operation during successive cycles of operation with differet samples. Also, with continued use, the spring constant tends to vary somewhat, further affecting the furnace operation. Finally, the copper electrodes tend to wear and deform, further affecting the reproducibility and furnace performance.